INDEX

Explanations

significant turns in conversation, likely indicating emotional or pivotal moments in dialogue

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

VersionUID

-0.41

 colspan

-0.36

...

-0.35

 cubiertas

-0.35

Ull

-0.34

Bou

-0.34

creation

-0.33

 inter

-0.33

Hau

-0.33

 Scal

-0.33

POSITIVE LOGITS

 dAtA

0.70

 يتيمه

0.63

 betweenstory

0.62

:✨

0.62

 resourceCulture

0.59

الحياه

0.57

ArrowToggle

0.57

脚注の使い方

0.56

PerformLayout

0.55

AnimationsModule

0.53

Activations Density 0.475%

significant turns in conversation, likely indicating emotional or pivotal moments in dialogue

No Comments

No Known Activations